INSPIRE: Tradeoffs in the Thermodynamics of Computation: A New Paradigm for Biological Information-Processing

Santa Fe Institute, Santa Fe NM

Investigators

Abstract

This INSPIRE project is jointly funded by the Chemistry of Life Processes Program in the Division of Chemistry in the Directorate for Mathematical and Physical Sciences, the Physics of Living Systems and Computational Physics Programs in the Division of Physics in the Directorate for Mathematical and Physical Sciences, the Systems and Synthetic Biology Cluster in the Division of Molecular and Cellular Biosciences in the Directorate for Biological Sciences, the INSPIRE Program and the Office of International Science and Engineering in the Office of Integrative Activities. This award is funding Dr. David Wolpert from the Santa Fe Institute, Dr. Seth Lloyd from the Massachusetts Institute of Technology, and Dr. Sebastian Deffner from the University of Maryland, to exploit the powerful new tools of non-equilibrium statistical physics that analyzing the energy tradeoffs inherent in all computation. Energy (thermodynamic) costs of computational systems play a role in everything from human-engineered computers (which release heat energy equal to about 5% of the annual energy expenditure in the US) to biological systems (which must harvest enough energy from their environment to fulfill their needs to think and move). To investigate these energy costs, this project pursues three, synergistic research thrusts. The research analyzes the fundamental energy tradeoffs faced by intracellular biochemical networks. The project analyzes the energy tradeoffs that engineers face when designing new computer technologies. Finally, the research integrate the energy tradeoffs of computational processes into the theory of computation. This project allows graduate students and postdoctoral fellows to acquire the skills needed for expanding what is known about the energy requirements of computation. In addition, the workshops held under the auspices of this project are building intellectual bridges connecting the multiple scientific disciplines that involve computational systems. Such brides are crucial to the development of a broadly applicable theory of thermodynamics of computation. This research project is undertaken to quantitatively analyze the tradeoffs relating the minimal free energy requirements and dissipation of a computer on the one hand, to several high-level properties of that computer on the other. This analysis considers the speed of the computer; the number of hidden internal states the computer can use as buffers; the variability of the inputs to the computer and the degree and type of noise in the computer. In addition to theoretical aspects of these issues and their consequences for computation theory, this research into the thermodynamics of computation is conducted in several domains that include: chromatin computers, the computation performed during RNA folding, the computation performed by biochemical networks and post-Moore computers that exploit "hybrid computation" involving both quantum and classical components. It is expected that, in addition to providing major insight into the role of the thermodynamics of computation in all those domains, this project may lay the foundation for an overarching theory of thermodynamics of computation in general.